Scribd Logo
Upload

Welcome to Scribd!

  • 8 views

    Week8 Tutorial

    Uploaded by

    Thu Phương

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Week8 Tutorial

    Uploaded by

    Thu Phương
    8 views17 pages

    Original Title

    Week8_Tutorial

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    8 views17 pages

    Week8 Tutorial

    Uploaded by

    Thu Phương

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 17

    BIG DATA

    Week 8 - Tutorial
    April 2024
    MEASURING CONNECTIONS BETWEEN VARIABLES

    Depending on the type of variables that we have there are different option on how to measure
    connections between variables

    Testing for
    connections
    between
    variables

    Contingency
    Correlation
    tables

    Nominal/Ordinal
    Ordinal Interval or ratio
    variables

    Spearman Pearson Spearman


    correlation correlation correlation
    coefficient coefficient coefficient

    2
    Which option is NOT a method to measure connections
    between variables?
    A. Correlation
    B. Regression
    C. Histogram
    D. Chi-square
    How would you evaluate the strength of the connection
    between two variables using statistical methods?
    A. p-value
    B. R-squared value
    C. Regression coefficient
    D. Standard deviation
    Analyze the relationship between two continuous
    variables using which statistical measure?
    A. Mean
    B. Correlation coefficient
    C. Standard deviation
    D. Mode
    CORRELATION ANALYSIS – MEASURING
    THE RELATIONSHIP BETWEEN TWO VARIABLES

    • Spearman correlation uses data rank to measure monotonicity between ordinal or


    continuous variables.
    • Pearson correlation, R or r, detects linear relationships between quantitative data.

    Strong Moderate Weak Weak Moderate Strong


    Negative Negative Negative Positive Positive Positive
    CorrelationCorrelation Correlation Correlation CorrelationCorrelation

    −1 Negative correlation: 𝑟 ≤ 0
    0 Positive correlation: 𝑟 ≥ 0
    1

    6
    Which correlation method detects linear relationships
    between quantitative data?
    A. Spearman correlation
    B. Pearson correlation
    C. T-test
    D. ANOVA
    Which method would you choose to measure the
    relationship between two continuous variables with a
    non-linear pattern?
    A. Spearman correlation
    B. Pearson correlation
    C. Linear regression
    D. Chi-square test
    What type of relationships does the Pearson correlation
    detect?
    A. Monotonic relationships
    B. Non-linear relationships
    C. Linear relationships
    D. No relationships
    What is the primary difference between Spearman
    correlation and Pearson correlation?
    A. Spearman correlation uses data rank for ordinal or continuous variables, while
    Pearson correlation detects linear relationships between quantitative data.
    B. Spearman correlation measures linear relationships, while Pearson correlation
    measures monotonicity.
    C. Spearman correlation is used for qualitative data, while Pearson correlation is used
    for quantitative data.
    If the sample Pearson correlation coefficient between no of
    hours of watching Tiktok and Big Data final exam score is equal
    to – 0.85, what does this tell you ? (Choose one or more correct
    statements)
    A. No of hours of watching Tiktok and Big Data final exam score have a negative strong
    linear relationship.
    B. The sample Pearson correlation coefficient suggests the more hours you watch Tiktok,
    the lower Big Data exam score you will receive.
    C. Suppose that a simple linear regression is used to predict Big Data final exam score
    when using the no of hours of watching Tiktok. Then no of hours of watching Tiktok
    explains 72.25% the variability in the Big Data final exam score.
    D. All of the above answer.
    PURPOSE OF REGRESSION ANALYSIS

    The purpose of regression analysis is to


    analyze relationships among variables.

    Answer the question of how much y


    Forecast or predict the value of y
    changes with changes in each of the
    based on the values of the X's
    X’s
    COEFFICIENT OF DETERMINATION, 𝑅2

    • Using 𝑅2 , we can assess the overall fit of a model.


    • 𝑅2 measures the proportion of the total variability in the response variable
    explained by the model.
    • 0 ≤ 𝑅2 ≤ 1. The closer 𝑅2 is to 1, the better the explanatory power of the
    model.
    • Note also that 𝑅2 = 𝑟 2 for a simple linear model (only), where r is the
    sample Pearson correlation coefficient.
    • When conducting simple linear regression, if we had a choice of candidates
    to use as the explanatory variable then we would prefer the one which
    maximises 𝑅2 .
    When evaluating a regression analysis model, what does
    a high R-squared value indicate?
    A. A strong correlation between the variables
    B. Overfitting of the model
    C. No relationship between the variables
    D. Underfitting of the model
    In regression analysis, what is the coefficient of
    determination used for?
    A. To quantify the extent to which the dependent variable is predictable
    B. To determine the slope of the regression line
    C. To identify outliers in the data set
    D. To calculate the standard error of the estimate
    When would you use regression analysis over
    correlation analysis?
    A. When you want to determine the strength and direction of the relationship
    between variables
    B. When you want to predict one variable based on another variable
    C. When you want to calculate the correlation coefficient
    D. Regression analysis and correlation analysis serve different purposes
    Below is JASP output when fitting Age vs Prevexp of employee_dataset.
    Model Summary - Age
    Model R R² Adjusted R² RMSE

    H₀ 0.000 0.000 0.000 11.777

    H₁ 0.801 0.642 0.642 7.051


    Coefficients

    Model Unstandardized Standard Error Standardized t p

    H₀ (Intercept) 62.981 0.542 116.304 < .001

    H₁ (Intercept) 54.329 0.440 123.481 < .001

    prevexp 0.090 0.003 0.801 29.085 < .001

    1. What is the value of the sample Pearson correlation coefficient between age and prevexp ? Interpret the value.

    2. What is the value of the coefficient of determination between age and prevexp ? Interpret the value.

    3. Write down the line of regression of age vs prevexp. Hence calculate the prediction of Age when prevexp = 100.

    You might also like